From video to RF? A fact well known to music enthusiasts is that the most critical parts of a recording/reproduction system are those where a transition occurs between one medium and another. For audio, that starts with the microphone where aural signals are changed to electrical signals. Next is the recording device, whether a compact disk, tape, hard drive or record. Then, the information is recovered with a transition from electrical back to aural signals. The greatest chance for error occurs when the signal mode or type must be changed.
The same issues exist in television - especially now that we need to deal with digital. Stations are faced with handling signals that originate from varying sources and formats. Then the digital signal formatted into the 19.39Mb/s bitstream, as specified in SMPTE-310M, must be converted to an analog RF signal which, when amplified, can be transmitted. The other end of the chain is at the receiver where the RF signal must be detected, decoded and displayed. The station can do little about the receiver - that portion is receiving enough press at present - but some precautions can ensure that the transmitted signal is as good as possible.
Critical points As in sound systems, the conversion from digital information to an RF signal, on frequency and ready to be amplified, may well be the most critical part of the system. This conversion involves multiple steps. Often referred to as the transport to transmission layer conversion, it includes frame synchronization, data randomization, Reed-Solomon encoding, data interleaving, trellis coding and DTV sync insertion. After filtering, the result is an 8VSB signal. That signal is then upconverted, usually in two steps, to the output signal on the desired channel.
All of that would be difficult enough just by itself. However, another critical problem must also be addressed in the exciter. The amplifiers, transmitter output filtering, diplexers (if used), transmission line system and antenna all add their little components of distortion, which can appear as a reduction in the transmitted signal-to-noise ratio. This in turn contributes to the bit error rate. The exciter must compensate for these errors. The most significant of the errors is probably nonlinear characteristics in the output device, whether one or more IOTs or a large number of solid state devices.
Different manufacturers handle these problems in different ways. For example, some manufacturers use a single exciter to drive multiple output devices with the correction circuitry adjusted to the best compromise settings. Others use separate exciters for each output device, or pair of output devices in large transmitters, which does allow for a little better fit of correction to error. Obviously, for solid state transmitters, the compromise solution is the only one possible.
The extent of the correction varies between manufacturers. Some take an RF sample from the transmission line after all filtering and combining is completed and do their final measurements at that point. Some also allow the predicted distortion from the line and antenna to be added to the sample signal, which should result in the greatest possible optimization of the system.
This may seem to be old hat and stale information to many readers. Rest assured that many transmitter technicians really don't yet comprehend the complexity of the process that takes place in the DTV exciter. It is significantly more complex than the more familiar NTSC analog exciter. Pre-correction is not only desirable, it is absolutely necessary in the digital world. It doesn't just concern how the picture looks in the field - it now has a significant effect on how many viewers will receive the signal. DTV receivers mute when the bit error rate becomes excessive. That is largely a matter of field strength, but it also depends on the characteristics of the transmitted signal, which, in turn are heavily dependent upon the correction circuitry. Another little item to consider is that correction isn't simply a matter of a capacitor here and there. The digital correction process is far more complex and demanding.
A brave new world? What does all this lead to for the station engineer? It means that the selection of a transmitter - and its attendant exciter - is far more complicated than in the days of analog. The old considerations still apply, including such items as power consumption, output device type, and operating cost and total price. But other considerations may be even more significant. For example, just how comprehensive and successful is the correction circuitry in reducing output errors? Equally important, how completely does the system provide the information necessary for correction adjustments? Picking the best system is far, far beyond the scope of this simple column. This is a decision to be made by the individual station or group engineers after discussions with the appropriate manufacturers. However, there is one major area that should be of concern to every engineer where this column's input is proper.
It is obvious that the modern DTV exciter cannot possibly be properly adjusted, operated or repaired without a complete understanding of its circuitry and the theory of its operation. That is not going to come from a cursory review of the instruction books. Stations should be hesitant to purchase any transmitter for which the manufacturer does not provide an in-depth training program. This should be done at the factory where the equipment exists to permit hands-on experience without screwing up the station's equipment. Experienced engineers who know the equipment in depth should lead the training courses. In addition, the program should be ongoing to allow new engineers to attend the school should the existing staff move on to greener pastures. The cost of such training should be established and considered in the selection of the transmitter supplier.
Station management must accept the fact that such training is not just desirable, it is absolutely imperative. Station staff cannot be expected to move into a totally new area with completely different demands without formal training in an appropriate venue. They also should not be faced with simultaneous demands to fix the ENG truck and repair the camera cable.
As for your author, after careful consideration of the demands of the digital world, he is giving careful consideration to returning to the world of glowing mercury vapor rectifiers and big glowing tubes. The faults were simpler, burnt-out filaments were the most common cause of trouble and no one had ever heard of bit error rate.